US20050181408A1

US20050181408A1 - Genetic analysis by sequence-specific sorting

Info

Publication number: US20050181408A1
Application number: US11/055,187
Authority: US
Inventors: Sydney Brenner
Original assignee: Individual
Current assignee: Agency for Science Technology and Research Singapore
Priority date: 2004-02-12
Filing date: 2005-02-10
Publication date: 2005-08-18
Also published as: DE602005018166D1; CA2555377A1; WO2005080604A2; WO2005080604A3; US7217522B2; EP1713936A4; EP1713936B1; EP1713936A2; AU2005214329A1; JP2007521833A; ATE451478T1

Abstract

The invention provides methods for sorting polynucleotides from a population based on predetermined sequence characteristics. In one aspect, the method of the invention is carried out by extending a primer annealed polynucleotides having predetermined sequence characteristics to incorporate a predetermined terminator having a capture moiety, capturing polynucleotides having extended primers by a capture agent that specifically binds to the capture moiety, and melting the captured polynucleotides from the extended primers to form a subpopulation of polynucleotides having the predetermined sequence characteristics. In another aspect, the method of the invention is carried out on a population of tagged polynucleotides so that after a subpopulation is selected, the members of the subpopulation may be simultaneously analyzed using the unique tags on the polynucleotides to convey analytical information to a hybridization array for a readout.

Description

This application claims priority from U.S. provisional application Ser. No. 60/543,887 filed 12 Feb. 2004, which is incorporated herein by reference in its entirety.

FIELD OF THE INVENTION

The invention relates generally to methods and compositions for analyzing complex populations of polynucleotides, and more particularly, to methods and compositions for partitioning a population of polynucleotides into one or more subpopulations of lesser complexity.

BACKGROUND

A major goal in genetics research is to understand how sequence variations in the genome relate to complex traits, particularly susceptibilities for common diseases such as diabetes, cancer, hypertension, and the like, e.g. Collins et al, Nature, 422: 835-847 (2003). The draft sequence of the human genome has provided a highly useful reference for assessing variation, but it is only a first step towards understanding how the estimated 10 million or more common single nucleotide polymorphisms (SNPs), and other polymorphisms, such as inversions, deletions, insertions, and the like, determine or affect states of health and disease. Many powerful analytical approaches have been developed to address this problem, but none appear to have adequate throughput or flexibility for the types of studies required to associate traits practically and reliably with genomic variation, e.g. Syvanen, Nature Reviews Genetics, 2: 930-942 (2001). For example, it would be desirable to carry out trait-association studies in which a large set of genetic markers from populations of affected and unaffected individuals are compared. Such studies depend on the non-random segregation, or linkage disequilibrium, between the genetic markers and genes involved in the trait or disease being studied. Unfortunately, the extent and distribution of linkage disequilibrium between regions of the human genome is not well understood, but it is currently believed that successful trait-association studies in humans would require the measurement of 30-50,000 markers per individual in populations of at least 300-400 affected individuals and an equal number of controls, Kruglyak and Nickerson, Nature Genetics, 27: 234-236 (2001); Lai, Genome Research, 11: 927-929 (2001); Risch and Merikangas, Science, 273: 1516-1517 (1996); Cardon and Bell, Nature Reviews Genetics, 2: 91-99 (2001).
One approach to dealing with such whole-genome studies is to create subsets of genomic DNA having reduced complexity with respect to the genomes being analyzed in order to simplify the analysis, e.g. Lisitsyn et al, Science, 259: 946-951 (1993); Vos et al, Nucleic Acids Research, 23: 4407-4414 (1995); Dong et al, Genome Research, 11: 1418-1424 (2001); Jordan et al, Proc. Natl. Acad. Sci., 99: 2942-2947 (2002); Weissman et al, U.S. Pat. No. 6,506,562; Sibson, U.S. Pat. No. 5,728,524; Degau et al, U.S. Pat. No. 5,858,656. Unfortunately, most of these techniques rely on some form of subtraction, sequence destruction, or direct or indirect size selection to create subsets, which are difficult to implement and reduced sensitivity.
In view of the above, the field of genetic analysis would be advanced by the availability of a method for converting a highly complex population of DNA, such as a mixture of genomes, into subsets having reduced complexity without requiring subtraction, or other sequence destroying, steps.

SUMMARY OF THE INVENTION

The invention provides methods and compositions for sorting polynucleotides from a population based on predetermined sequence characteristics. In one aspect, the method of the invention is carried out by the following steps: (i) extending a primer annealed polynucleotides having predetermined sequence characteristics to incorporate a predetermined terminator having a capture moiety, (ii) capturing polynucleotides having extended primers by a capture agent that specifically binds to the capture moiety, and (iii) melting the captured polynucleotides from the extended primers to form a subpopulation of polynucleotides having the predetermined sequence characteristics.
In another aspect, the population of polynucleotides comprises fragments from a population of genomes, wherein the fragments from each genome has the same unique oligonucleotide tag attached. In this aspect, the invention includes a method of determining a frequency of a nucleotide at a predetermined locus in a population of genomes, such method comprising the following steps: (i) separately generating fragments of each genome of the population; (ii) attaching a unique oligonucleotide tag to each genome; (iii) selecting fragments from each genome that contains the predetermined locus; (iv) generating a labeled oligonucleotide tag from each unique oligonucleotide tag, the labeled oligonucleotide tag generating a signal indicative of the nucleotide at the predetermined locus; and (v) determining the frequency of the nucleotide at the predetermined locus by detecting the signals generated by the labeled oligonucleotide tags specifically hybridized with their respective tag complements, the respective tag complements being attached in spatially discrete regions on the one or more solid phase supports.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGS. 1A-1F illustrate the selection of particular fragments by common sequence elements.
FIGS. 2A-2D illustrate the application of the invention for selecting particular haplotypes.
FIGS. 3A-3D illustrate hybridization tags with “commas” and a hybridization tag with the “comma-less” property.
FIG. 4 lists melting temperatures of selected tags consisting of four words each having the comma-less property.

DEFINITIONS

“Addressable” in reference to tag complements means that the nucleotide sequence, or perhaps other physical or chemical characteristics, of a tag complement can be determined from its address, i.e. a one-to-one correspondence between the sequence or other property of the tag complement and a spatial location on, or characteristic of, the solid phase support to which it is attached. Preferably, an address of a tag complement is a spatial location, e.g. the planar coordinates of a particular region containing copies of the tag complement. However, tag complements may be addressed in other ways too, e.g. by microparticle size, shape, color, frequency of micro-transponder, or the like, e.g. Chandler et al, PCT publication WO 97/14028.
“Allele frequency” in reference to a genetic locus, a sequence marker, or the site of a nucleotide means the frequency of occurrence of a sequence or nucleotide at such genetic loci or the frequency of occurrence of such sequence marker, with respect to a population of individuals. In some contexts, an allele frequency may also refer to the frequency of sequences not identical to, or exactly complementary to, a reference sequence.
“Amplicon” means the product of an amplification reaction. That is, it is a population of polynucleotides, usually double stranded, that are replicated from one or more starting sequences. The one or more starting sequences may be one or more copies of the same sequence, or it may be a mixture of different sequences. Amplicons may be produced in a polymerase chain reaction (PCR), by replication in a cloning vector, or by linear amplification by an RNA polymerase, such as T7 or SP6, or by like techniques.
“Analyte” means any molecule, including organic, inorganic, or biomolecule, whose presence or absence or quantity or concentration in a sample is to be determined in an assay. In particular, biomolecule analytes include oligonucleotides, polynucleotides, genomic fragments, messenger RNAs (mRNAs), proteins, antibodies, enzymes, complementary DNAs (cDNAs), and like compounds.
“Complement” or “tag complement” as used herein in reference to oligonucleotide tags refers to an oligonucleotide to which an oligonucleotide tag specifically hybridizes to form a perfectly matched duplex or triplex. In embodiments where specific hybridization results in a triplex, the oligonucleotide tag may be selected to be either double stranded or single stranded. Thus, where triplexes are formed, the term “complement” is meant to encompass either a double stranded complement of a single stranded oligonucleotide tag or a single stranded complement of a double stranded oligonucleotide tag.

“Complexity” in reference to a population of double stranded or single stranded polynucleotides means the number of different species of polynucleotide present in the population. The related concept, “kinetic complexity” in reference to genomic DNA means the total number of basepairs present in non-repeating sequences, e.g. Wetmur, Critical Reviews in Biochemistry and Molecular Biology, 26: 227-259 (1991); Britten and Davidson, chapter 1 in Hames et al, editors, Nucleic Acid Hybridization: A Practical Approach (IRL Press, Oxford, 1985). For example, the following populations have the indicated sizes and complexities:



	Population	Com-
Population	Size	plexity

agtctactggtttca	3	3
tcagatgaccaaagt
(SEQ ID NO: 1)

gggttggggtttacccctttagc
cccaaccccaaatggggaaatcg
(SEQ ID NO: 2)

tattagcttacttggcctta
ataatcgaatgaaccggaat
(SEQ ID NO: 3)

agtctactggtttcaattaattaatt	2	2
tcagatgaccaaagttaattaattaa
(SEQ ID NO: 4)

gggttggggtttacccctttagc
cccaaccccaaatggggaaatcg
(SEQ ID NO: 2)

gggttggggtttacccctttagc	5	3
(SEQ ID NO: 5)

tcagatgaccaaagt
(SEQ ID NO: 6)

tcagatgaccaaagt
(SEQ ID NO: 6)

tcagatgaccaaagt
(SEQ ID NO: 6)

tcagatgaccaaagttcagatgaccaaagt
(SEQ ID NO: 7)

cccttagctg	agggct	8	3
(SEQ ID NO: 8)

cccttagctg	agggct
(SEQ ID NO: 8)

cccttagctg	agggct
(SEQ ID NO: 8)

cccttagctg	agggctc
(SEQ ID NO: 8)

In each set, all four differ in both positions from all the other members of the set, but when the four different sets are compared with each other, one base is held in common with one member of the other set. For example, in set I, eight different words can be created by combining doublets from set I with doublets from set II in the I-II order and the II-I order. Since each of these sets contain doublets that are the reverse complements of the other, the combinations are made such that none of I-II four-base words are the inverse complements of the II-I four-base words. Thus, if the I-II words are selected as follows: GTCT, TGTC, ACAG, and CAGA, then the II-I words can be defined only as follows:

AGCA or AGGT

GAAC GATG

CTTG CTAC

TCGT TCCA
an arrangement which conserves the constraint that the members of each set differs by three bases from any member of the same set. From the above sets, several sets of words for comma-less tags can be constructed. Taking the first two sets, an “A” to the end of each words of the first set, and a “T” to the end of each word of the second set to give the following:

AGCAT GTCTA

GAACT TGTCA

CTTGT ACAGA

TCGTT CAGAA
Although the same process does not work with sets II and IV above because in III the doublets are self-complementary, further sets of words can be created by switching the I-II into II-I and vice versa, and adding the bases as above, which gives:

CTGTA CAAGT

TCTGA ACGAT

AGACA TGCTT

GACAA GTTCT
For tags not used in enzymatic processing, such as anti-tags synthesized on a solid phase support, the following sets employing deoxyinosine may be employed:

AICAT GTCTA

GAACT TGTCA

CTTGT ACAGA

TCITT CAGAA

and

CTGTA CAAGT

TCTGA ACIAT

AGACA TICTT

GACAA GTTCT

Further sets of words for constructing comma-less tags are listed in FIG. 4.

Tag Complements Hybridization and Readout

Preferably, tag complements are synthesized on the surface of a solid phase support, such as a microscopic bead or a specific location on an array of synthesis locations on a single support, such that populations of identical, or substantially identical, sequences are produced in specific regions. That is, the surface of each support, in the case of a bead, or of each region, in the case of an array, is derivatized by copies of only one type of tag complement having a particular sequence. The population of such beads or regions contains a repertoire of tag complements each with distinct sequences. As used herein in reference to hybridization tags, tag complements, and synthesis tags, the term “repertoire” means the total number of different tags or tag complements in a given set or population.
Solid phase supports containing tag complements may take a variety of forms, e.g. particulate, single-piece and planar, such as a glass slide, and may be composed of a variety of materials, e.g. glass, plastic, silicon, polystyrene, or the like. Particulate solid phase supports include microspheres, particularly fluorescently labeled microspheres, e.g. Han et al, Nature Biotechnology, 19: 631-635 (2001); Kettman et al, Cytometry, 33: 234-243 (1998); and the like. Preferably, hybridization tags are detected by hybridizing them to their complementary sequences on a conventional microarray. Such microarrays may be manufactured by several alternative techniques, such as photo-lithographic optical methods, e.g. Pirrung et al, U.S. Pat. No. 5,143,854, Fodor et al, U.S. Pat. Nos. 5,800,992; 5,445,934; and 5,744,305; fluid channel-delivery methods, e.g. Southern et al, Nucleic Acids Research, 20: 1675-1678 and 1679-1684 (1992); Matson et al, U.S. Pat. No. 5,429,807, and Coassin et al, U.S. Pat. Nos. 5,583,211 and 5,554,501; spotting methods using functionalized oligonucleotides, e.g. Ghosh et al, U.S. Pat. No. 5,663,242; and Bahl et al, U.S. Pat. No. 5,215,882; droplet delivery methods, e.g. Caren et al, U.S. Pat. No. 6,323,043; Hughes et al, Nature Biotechnology, 19: 342-347 (2001); and the like. The above patents disclosing the synthesis of spatially addressable microarrays of oligonucleotides are hereby incorporated by reference. Microarrays used with the invention contain from 50 to 500,000 hybridization sites; or from 100 to 250,000 hybridization sites; or from 100 to 40,000 hybridization sites; and preferably, they contain from 100 to 32,000 hybridization sites; or from 100 to 20,000 hybridization sites; or from 100 to 10,000 hybridization sites.
Guidance for selecting conditions and materials for applying labeled oligonucleotide probes to microarrays may be found in the literature, e.g. Wetmur, Crit. Rev. Biochem. Mol. Biol., 26: 227-259 (1991); DeRisi et al, Science, 278: 680-686 (1997); Wang et al, Science, 280: 1077-1082 (1998); Duggan et al, Nature Genetics, 21: 10-14 (1999); Schena, Editor, Microarrays: A Practical Approach (IRL Press, Washington, 2000); Hughes et al (cited above); Fan et al, Genomics Research, 10: 853-860 (2000); and like references. These references are hereby incorporated by reference. Typically, application of hybridization tags to a solid phase support includes three steps: treatment with a pre-hybridization buffer, treatment with a hybridization buffer that includes the probes, and washing under stringent conditions. A pre-hybridization step is employed to suppress potential sites for non-specific binding of probe. Preferably, pre-hybridization and hybridization buffers have a salt concentration of between about 0.8-1.2 M and a pH between about 7.0 and 8.3. Preferably, a pre-hybridization buffer comprises one or more blocking agents such as Denhardt's solution, heparin, fragmented denature salmon sperm DNA, bovine serum albumin (BSA), SDS or other detergent, and the like. An exemplary pre-hybridization buffer comprises 6×SSC (or 6×SSPE), 5× Denhardt's solution, 0.5% SDS, and 100 μg/ml denatured, fragmented salmon sperm DNA, or an equivalent defined-sequence nucleic acid. Another exemplary pre-hybridization buffer comprises 6×-SSPE-T (0.9 M NaCl, 60 mM NaH2PO4, 6 mM EDTA (pH 7.4), 0.005% Triton X-100) and 0.5 mg/ml BSA. Pre-hybridization and hybridization buffers may also contain organic solvents, such as formamide to control stringency, tetramethylammonium chloride to negate base-specific effects, and the like. An exemplary hybridization buffer is SSPE-T and the desired concentration of isostringency probe. After hybridization, unbound and non-specifically bound isostringency probe is removed by washing the detection support under stringent conditions. Preferably, stringency of the wash solution is controlled by temperature, organic solvent concentration, or salt concentration. More preferably, the stringency of the wash conditions are determined to be about 2-5° C. below the melting temperature of the isostringency probes at the salt concentration and pH of the wash solution. Preferably, the salt concentration of the wash solution is between about 0.01 to 0.1 M.
Instruments for measuring optical signals, especially fluorescent signals, from labeled tags hybridized to targets on a microarray are described in the following references which are incorporated by reference: Stern et al, PCT publication WO 95/22058; Resnick et al, U.S. Pat. No. 4,125,828; Kamaukhov et al, U.S. Pat. No. ,354,114; Trulson et al, U.S. Pat. No. 5,578,832; Pallas et al, PCT publication WO 98/53300; Brenner et al, Nature Biotechnology, 18: 630-634 (2000); and the like.
When tag complements are attached to or synthesized on microbeads, a wide variety of solid phase materials may be used with the invention, including microbeads made of controlled pore glass (CPG), highly cross-linked polystyrene, acrylic copolymers, cellulose, nylon, dextran, latex, polyacrolein, and the like, disclosed in the following exemplary references: Meth. Enzymol., Section A, pages 11-147, vol. 44 (Academic Press, New York, 1976); U.S. Pat. Nos. 4,678,814; 4,413,070; and 4,046;720; and Pon, Chapter 19, in Agrawal, editor, Methods in Molecular Biology, Vol. 20, (Humana Press, Totowa, N.J., 1993). Microbead supports further include commercially available nucleoside-derivatized CPG and polystyrene beads (e.g. available from Applied Biosystems, Foster City, Calif.); derivatized magnetic beads; polystyrene grafted with polyethylene glycol (e.g., TentaGel™, Rapp Polymere, Tubingen Germany); and the like. Generally, the size and shape of a microbead is not critical; however, microbeads in the size range of a few, e.g. 1-2, to several hundred, e.g. 200-1000 μm diameter are preferable, as they facilitate the construction and manipulation of large repertoires of oligonucleotide tags with minimal reagent and sample usage. Preferably, glycidal methacrylate (GMA) beads available from Bangs Laboratories (Carmel, Ind.) are used as microbeads in the invention. Such microbeads are useful in a variety of sizes and are available with a variety of linkage groups for synthesizing tags and/or tag complements.

Hybridization Code

In one aspect, hybridization codes of the invention consist of five bases and are assembled into hybridization tags following a procedure similar to that described in Brenner and Williams (cited above). Using synthesis tags, hybridization tags are constructed that are complements of the anti-tags attached to solid phase supports, such as microarrays. Such tags have the following form (SEQ ID NO: 9):

. . . GCATCNNNNN-H₁-H₂-NNNNNNNNCATCC . . . (I)

Sfa NI Fok I

where H₁and H₂are words of a hybridization tag as described above, for example 4-mer words. Such words may vary in length depending on the embodiment, but generally are in the range of from 2 to 10 nucleotides in length; or they may be in the range of from 3 to 6 nucleotides in length. One factor in selecting word length is whether they are processed by restriction enzymes, such as type IIs restriction enzymes, whose recognition and cleavage characteristics may dictate word length. Using an eight-word set described above, 64 such di-words are constructed, cloned in conventional vectors, and the DNA can be obtained thereafter by PCR. These reagents containing pairs of hybridization “words” are used to form word-pair conversion adaptors, described more fully below.
The principle of successively adding words is as follows. Assuming a word is in place and that a successive word is to be added. Since the previous word can be any of the eight words, then the material to be added will need to have all possibilities in the next position, call this “H₂”, and there would be eight such sets. Thus, when the Sfa NI site is cut we will have the following end:

pZ₁Z₁Z₁Z₁Z₁Z₀Z₀Z₀Z₀Z₀. . . (II)

Z₁Z₀Z₀Z₀Z₀Z₀. . .
where the “Z₁'s” are the nucleotides of the added word, the “Z₀'s” are the nucleotides of the previous word, and “p” is a phosphate group. The new word is added by cutting the di-words of formula (I) at the Fok I site to give (SEQ ID NO: 10):

. . . GCATCNNNNN-Z₂Z₂Z₂Z₂Z₂Z₂

. . . CGTAGNNNNN-Z₂Z₂Z₂Z₂Z₂Z₂Z_XZ_XZ_XZ_Xp

where the “Z₂'s” are the nucleotides of the next word, and the “Z_x's” are the nucleotides of all the possible cleavage products. The cleavage product includes ends complementary to all of the possible ends of the cleavage product of formula (II). Thus, ligation of the two products permits combinatorial synthesis of the tags.

Tagging Polynucleotides

In one aspect of the invention, all fragments of each genome of a population of genomes are labeled with one combination of words selected from a repertoire of 32,768 (=8⁵) five-word oligonucleotide tags. Once each genome has a unique tag, then common-sequence fragments, e.g. a restriction fragment from a particular locus, can be selected using the method of the invention. The tags may then be used to convey information about the fragments, e.g. the identity of a nucleotide at a particular locus, to a hybridization array for a readout. One of ordinary skill in the art understands that the selection of 5-word oligonucleotide tags of five nucleotides each and the use of commaless tags are design choices that may be varied depending on the goals and constraints of any particular application. In one embodiment the following eight-word minimally cross-hydridizing set may be used to construct the above repertoire. As described below, preferably, each word is cloned in a plasmid with additional elements for aiding in the construction of oligonucleotide tags.

AGCAT GTCTA

GAACT TGACA

TCTGT ACGAA

CTGTT CATCA

Using these words, 64 di-words are prepared in separate plasmids as described in Brenner and Williams (cited above), which is incorporated by reference.
A. Single-Word Library and Counting Array Element.
In one embodiment, the single word library contains a ten-base sequence [G/T; G/T; A/T]₃G/T, where “x/T” is an equal mixture of the two bases “x” and “T” at a particular locus. This element encodes a repertoire of 1024 (=2¹⁰) different sequences that permits sequences to be counted by hybridization of copies of the sequence to an array of complementary sequences, i.e. a “counting” array. This element is referred to herein as the “Counting Array” or “CAR” element. In this embodiment, about 30 copies of each genome are tagged and each is labeled with one unique sequence. Thus, if any sorted molecule is found to have a unique sequence for this array, it is not a genome difference that should have multiple sequences, and is likely to represent an error in the process which has resulted in an altered molecule. Note that however much any fragment is amplified that it will always possess the original sequences in the counting array, preserving cardinality as distinct from the concentration of DNA.
A plasmid having the following characteristics is constructed: (i) no SapI site, and (ii) a sequence of restriction sites:

GGGCCC . . . AGGCCT . . . GGTACC

(ApaI) (BspE1) (KpnI)
These sites each have “GG” which is absent from tags constructed from the words of the above set. Next for each word the strands of following element are synthesized (SEQ ID NO: 11):

5′-pCNNNNNNNNNNGCATCNNNNN [WORD] A

3′-CCGGGNNNNNNNNNNCGTAGNNNNN [WORD] TCCGGp

(Sfa N1)

where lower case “p” represents a phosphate group. After annealing the strands, the element is cloned into the above plasmid by cleaving with ApaI and Bsp E1. Several plasmids are picked for each word and the clones are sequenced to check the accuracy of the sequence, after which one is selected for use in tag construction. Elements for the “counting” array are synthesized and also a second primer binding site which will be required for later amplification. After synthesis, the following structure is obtained (SEQ ID NO: 12):


3′-NNNTCCGGA [N₁₅] CCCTG [(G/T; G/T; A/T) ₃ G/T]
BspE1 BsmF1 CAR element
GTTGCTTCTCGCCATGGNNNN
SapI KpnI

Using the primer “5′-NNNAGGCCT[N₁₅]GGGAC” (SEQ ID NO: 13) the above is copied, cleaved with KpnI and BspE1, and cloned into each of the single-word plasmids. 10⁴clones of each are isolated to make sure that all the sequences of the counting array are in the library.

This embodiment is designed to attach tags to fragments generated by cleaving with the “↓GATC” family of restriction endonucleases. These enzymes permit the generation of the fragments of several different lengths:

Average

Enzyme Recognition Site Fragment Length

Bam HI G↓GATCC 4 Kb

Bam HI + BglII G↓GATCC + G↓GATCT 2 Kb

Bst YI R↓GATCY 1 Kb

Sau 3a ↓GATC 256 bp
All of these leave the same end when cleaved, namely:

5′-NN

NNCTAGp
where “p” is a phosphate group. This may be filled in with a single dGTP to give a three-base overhang:

5′-NNG

NNCTAGp
After such filling, polynucleotides or cloning vectors cut with SapI (underlined below), which leaves the following ends:

5′- . . . NN GATCGAAGAGC . . .

. . . NNTAGp GCTTCTCG . . .

permits efficient and directional cloning of fragments.

The final construct has the following structure:


. . . [ApaI site]N ₁₀ [SfaN1 site]N ₅ [word][BspE1
Primer X
site]N ₁₅ [BsmF1 site][CAR][SapI site][KpnI
Primer Y Primer Z
site] . . .

were “N” are arbitrarily selected nucleotides and “CAR” is a counting array element, as described above.
B. Double-Word Libraries.

Here a library of 64 vectors is disclosed each containing one of the 64 possible two-word, or “di-word,” concatenations of words from the 8-word library flanked by primer binding sites. This double-word library is then used essentially as described in Brenner and Williams (cited above) to construct oligonucleotide tags. In this embodiment, the first flanking primer binding site is that shown above as “Primer X,” and the other contains a recognition site for FokI, 5′-GGATG(9/13), which contains “GG” and therefore cannot cut any of the words described above.
The following vector elements are synthesized (SEQ ID NO: 14):

5′-pCN₁₀[SfaN1 site] N₅[word 1] [word 2] N₈CATCC
and (SEQ ID NO: 15):

3′-CCGGGN₁₀[SfaN1 site]N₅[word 1] [word 2]

N₉GTAGGCTAG

where it is understood that the “word 1” and “word 2” refer to both word sequences and their respective complements. After annealing the above fragments to form a doublestranded element, it is cloned into a plasmid digested with ApaI and BamHI. To assure the accuracy of the incorporation, several clones of each “double word” vector are selected and sequenced. Copies of di-words may be conveniently obtained by PRC using a biotinylated X primer and another primer.
C. Tagging Genomes.
About 1 ng of human DNA (about 30 copies of the haploid genome) is digested with Bst Y1 to give fragments of an average size of 1 Kb, after which ends are filled in with dGTP to give 3-base ends as described above.

The eight single word libraries, labeled A-H, are amplified and cut with SapI to generate the following single-word fragment:


[ApaI site]N ₁₀ [SfaN1 site]N ₅ [word][BspE1 site]N ₁₅
[BsmF1 site][CAR]

[ApaI site]N ₁₀ [SfaN1 site]N ₅ [word][BspE1 site]N ₁₅
Primer X Primer Y.
[BsmF1 site][CAR]CTAp

64 genomes are tagged in one batch as follows. 64 reaction vessels are arranged in an 8×8 array wherein each row, 1-8, contains 8 vessels labeled A-H. To each vessel a different Bst YI-digested genome is added, after which a different single-word fragment, A-H, is added to vessels 1-8, in each row to give the following array of reaction vessels with the following single-word fragments:

Row Single-Word Fragment

1 A B C D E F G H

2 A B C D E F G H

3 A B C D E F G H

4 A B C D E F G H

5 A B C D E F G H

6 A B C D E F G H

7 A B C D E F G H

8 A B C D E F G H

The single-word fragments are ligated to the genome fragments to give genome fragments having single-word fragments on both ends. These fragments are processed as follows so that a single-word fragment is on only one end. First, the reaction constituents from every vessel in each row are pooled so that eight mixed samples are obtained.



	Row	Single-Word Fragment

	1	A + B + C + D + E + F + G + H
	2	A + B + C + D + E + F + G + H
	3	A + B + C + D + E + F + G + H
	4	A + B + C + D + E + F + G + H
	5	A + B + C + D + E + F + G + H
	6	A + B + C + D + E + F + G + H
	7	A + B + C + D + E + F + G + H
	8	A + B + C + D + E + F + G + H

The DNA of each of the eight vessels is denatured and Primer Y (pAGGCCTN₁₅GGGAC) (SEQ ID NO: 16) is added to prime the 3′ tag sequence of each of the single strands as follows (SEQ ID NO: 17 AND SEQ ID NOL 18):

AGGCCTN₁₅GGGAC

TCCGGAN₁₅CCCTG [CAR]CTAG [fragment]CTAG [CAR]

GTCCC . . .

The primer is extended using 5-Me-dCTP to give the following (SEQ ID NO: 19 AND SEQ ID NO: 20):


AGGCCTN₁₅GGGAC	[frag-	[CAR]GTC(Me)C(Me)C(Me) . . .
[CAR]GATC(Me)	ment]
	GATC
	(Me)

TCCGGAN₁₅CCCTG	[frag-	[CAR]CAG G G . . .
[CAR]CTAG	ment]
	CTAG

All of the BsmF1 sites of the fragments are protected by half methylation, except for the site to the left of the tag. When the fragments are cleaved with BsmF1, the lefthand tag is removed up to the “GATC” site, leaving the following (SEQ ID NO: 21):


↓
. . . GGGAC [CAR]GATC
[fragment] . . .

. . . CCCTG [CAR]CTAG
[fragment] . . .
↑
↓
GATC[fragment]GATC [CAR] [BsmF1 site] [Primer Y]
[word]N₅[SfaN1 site] [Primer X]

[fragment]CTAG [CAR] [BsmF1 site] [Primer Y]
[word]N₅[SfaN1 site] [Primer X]

The “GATC” overhang is filled in with dGTP and ligated to the following adaptor containing a primer binding site for sequencing (SEQ ID NO: 22):

N₂₀GC^Me ATCAG

N₂₀CG TAGTCTAGp
The methylated C in the upper strand protects the lefthand site while the right hand portion of the fragments are manipulated. Words are added as follows. First, the C's of the bottom strand are replaced with 5-methyl-C's. This is accomplished by denaturing the above fragments, priming with a biotinylated Primer X (5′-biotin-GGGCCCN₁₀[Sfa N1 site]N₅), copying with 5-Me-CTP, and removing the strands with avidinated support. The fragments are released by cleaving with Sfa N1 to give in each of the eight vessels the sequences:

[fragment]GATC[CAR] [Primer Y]W

[fragment]CTAG[CAR] [Primer Y]WWWWWp

where all eight words are represented in the overhang and “W” represents a nucleotide of a word or its complement. Next the di-word libraries are pooled, cleaved with FokI, then ligated to the above fragment to add the next word. The process is continued until the desired number of words is added to the genomic fragments to complete the tags.

Claims

1. A method of sorting polynucleotides having predetermined sequence characteristics, the method comprising the steps of:

extending a primer annealed polynucleotides having predetermined sequence characteristics to incorporate a predetermined terminator having a capture moiety;

capturing polynucleotides having extended primers by a capture agent that specifically binds to the capture moiety; and

melting the captured polynucleotides from the extended primers.

2. A method of producing a subpopulation of polynucleotides having a complexity less than that of a parent population, the method comprising the steps of:

annealing a primer to polynucleotides of a parent population to form primer-polynucleotide duplexes;

extending the primer to incorporate a predetermined terminator having a capture moiety;

separating the primer-polynucleotide duplexes having an extended primer from the parent population by specifically binding the capture moiety of the predetermined terminator to a capture agent attached to a solid phase support;

melting the primer-polynucleotide duplexes to form a subpopulation of polynucleotides having a complexity less than that of the parent population.

3. A method of producing a population of polynucleotides having a desired complexity less than that of a parent population, the method comprising the steps of:

(a) annealing a primer to polynucleotides of a parent population to form primer-polynucleotide duplexes;

(b) extending the primer to incorporate a predetermined terminator having a capture moiety;

(c) separating the primer-polynucleotide duplexes having an extended primer from the parent population by specifically binding the capture moiety of the predetermined terminator to a capture agent attached to a solid phase support;

(d) melting the primer-polynucleotide duplexes to form a selected population of polynucleotides having a complexity less than that of the parent population, the selected population forming a parent population for subsequent steps;

(e) repeating steps (a) through (d) until a selected population of the desired complexity is obtained.

4. The method of claim 3 further comprising a step of replicating said selected population after said step of melting.

5. The method of claim 4 wherein during each said step of repeating steps (a) through (d), said primer anneals to a different primer binding site on said polynucleotides of said parent population or said selected population.

6. The method of claim 5 wherein in each successive step of repeating steps (a) through (d), said different primer binding site is shifted along said polynucleotides at least one nucleotide in a primer extension direction.

7. The method of claim 5 wherein in each successive step of repeating steps (a) through (d), said different primer binding site is at a different and non-overlapping locus of said polynucleotides.

8. The method of claim 7 wherein said different and non-overlapping locus is adjacent to and upstream of a single nucleotide polymorphism site.

9-10. (canceled)